前幾層的資料夾是有包含ROS的，純車尾燈辨識的程式在 itriadv/src/sensing/ itri_vehicle_signal/model 底下 執行前要先把dataset準備好

Rear Signal Dataset http://vllab1.ucmerced.edu/~hhsu22/rear_signal/rear_signal

If just want to use pre-train model to run the VGG16-WaveNet, go to itriadv/src/sensing/ itri_vehicle_signal/model, and run clasify.py

pre-train model (不知道學校的gmail甚麼時候會關起來) Turn light : https://drive.google.com/file/d/1LkLLo_ki_tr7zLdrQl2mT9PABS7qGdcY/view?usp=sharing Brake light : https://drive.google.com/file/d/1khSyjkEBZW-zH3pzMszdr8MGXV2JYVy1/view?usp=sharing

This repository contains the source codes of self-driving car maintained by ITRI ICL-U.

Prerequisite for ROS

1. Ubuntu 18.04
2. ROS Melodic
3. OpenCV 4.2.0
4. PCL 1.9.1
5. Cuda (Optional)
6. TensorRT (Optional)

See Automatic Scripts about how to install tem.

How to build

catkin_make -DCMAKE_BUILD_TYPE=Release

Tips: Run sudo apt-get install ccache to speed up the build process.

Developers Guide

1. How to commit code
2. Coding style